I have a task of finding confidence interval for difference of two samples means (its ARPU - avarage revenue per user). I need to find confidence interval for (Mean1 - Mean2). My guess is to use bootstrap in this task, but I'm not sure how to calculate standart deviation and confidence interval for difference.
What formulas should I use to get correct statistics? First sample size 2 800 000 and its mean 754. Second sample size 115 000 and its mean 711.
I can make bootstrap for 100 times for first sample and 1000 times for the second (I don't have memory for more) and using R.